Clément Pinard
Senior Applied Scientist at Onfido
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolClément Pinard is a Senior Applied Scientist based in Paris with 11 years of experience bridging academic research and production-grade computer vision. He holds a PhD from ENSTA Paris and has applied deep learning to real-world problems across companies like Onfido, XXII and Contentsquare, focusing recently on document fraud and object detection. His background includes novel work on monocular depth estimation for obstacle avoidance during a CIFRE PhD at Parrot, and contributions to well-known open-source projects such as a PyTorch port of SfmLearner where he improved data pipelines, normalization, and depth/pose validation. Comfortable moving between research and engineering, he has a track record of shipping robust vision models and applying geometric methods in neural networks. Colleagues rely on him for turning advanced algorithms into production-ready systems that solve concrete safety and fraud-detection problems.
11 years of coding experience
5 years of employment as a software developer
Systèmes électroniques, réseaux et images, Systèmes électroniques, réseaux et images at École Supérieure d'ÉlectricitéClasse Preparatoire aux grandes Ecoles (MP*), Classe Preparatoire aux grandes Ecoles (MP*) at College stanislasDoctor of Philosophy - PhD, Laboratoire U2IS, Doctor of Philosophy - PhD, Laboratoire U2IS at ENSTA ParisClasse Preparatoire aux grandes Ecoles (MPSI), Classe Preparatoire aux grandes Ecoles (MPSI) at Lycée Alain Fournier
French, English, Spanish
Stackoverflow
96reputation
5kreached
1answer
1question
Github Skills (15)
data-preprocessing10
depth-estimation10
data-loading10
pytorch10
computer-vision10
unsupervised-learning10
dataprep10
preprocessing10
python10
load-data10
preprocess10
image-processing9
machine-learning9
pandas6
tqdm6
Programming languages (9)
Github contributions (5)
ClementPinard/SfmLearner-Pytorch
Nov 2017 - Jan 2023
Pytorch version of SfmLearner from Tinghui Zhou et al.
Role in this project:
ML Engineer
ML EngineerContributions:27 commits, 17 PRs, 85 pushes in 5 years 3 months
Contributions summary:Clément primarily contributed to the project by modifying and adding functionality related to the data loading and processing pipelines within the SfMlearner-pytorch repository. They fixed issues in data loading, applied the correct normalization in inference, and added a new shifted sequence folder. Moreover, the user worked on adding functionalities related to depth and pose estimations, and made improvements to the validation visualizations.
pytorchkittidisparitydepthunsupervised
ClementPinard/FlowNetTorch
Dec 2016 - Apr 2017
Torch implementation of Fischer et al. FlowNet training code
Contributions:9 commits, 29 pushes, 1 branch in 3 months
torchflownet
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.